CN106600574A - Landslide extraction method based on remote-sensing image and altitude data - Google Patents
Landslide extraction method based on remote-sensing image and altitude data Download PDFInfo
- Publication number
- CN106600574A CN106600574A CN201610718410.9A CN201610718410A CN106600574A CN 106600574 A CN106600574 A CN 106600574A CN 201610718410 A CN201610718410 A CN 201610718410A CN 106600574 A CN106600574 A CN 106600574A
- Authority
- CN
- China
- Prior art keywords
- landslide
- image
- region
- altitude data
- remote sensing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20036—Morphological image processing
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a landslide extraction method based on a large-range remote-sensing image and altitude data. According to the method, on the basis of a remote-sensing image of a study area and corresponding altitude data, a bare soil region in the image is enhanced by using a significance concept to obtain a significance probability graph, wherein each pixel belongs to a probability graph of a landslide. On the basis of an expansion operation in a morphological algorithm, little large-plaque bare soil in the image is communicated to form a large area and thus the difference with a small landslide area is enhanced, so that the non-landslide bare soil area can be rejected to obtain a landslide potential region. On the basis of the characteristic of frequent occurrence of the landslide at a mountainous area, the potential landslide region at the mountain slope is kept by combining the altitude data to obtain a final landslide extraction result. With the method disclosed by the invention, problems of small research range and simple situation for the existing landslide extraction example can be solved; a technological base is provided for large-range practical rapid landslide extraction; and thus the method plays an important role in post-disaster emergency response and rapid landslide area positioning.
Description
Technical field：
It is a kind of based on remote sensing image on a large scale and altitude data the present invention relates to image procossing, area of pattern recognition
Landslide extracting method.
Background technology：
Landslide, as one of main natural disaster, Jing often serious to the life and composition of estate of mankind threat.In mistake
The decades gone, the landslide for frequently occurring has caused the very big concern of society.Fast and accurately detection landslide not only helps
The mechanism that landslide occurs is understood in people, can be more offer guidance data of taking emergency measures after calamity, be disaster-stricken
Scale evaluation provides reliable foundation.
The continuous development of satellite sensor and remotely-sensed data resolution it is constantly improve so that on a large scale landslide monitoring becomes
May.At present, being based on change-detection the method for landslide detection more, judging to slide by contrasting multiple phase images in same research area
The generation on slope.Wherein, normalized differential vegetation index NDVI (Normalized Difference Vegetation Index) is commonly used
Vegetation information in strengthen image, and then landslide is distinguished from vegetation.Other spectral indexes and post-classification comparison side
Method is also commonly used to extract landslide, especially for multiband remote sensing image.Application of the Object--oriented method in landslide is extracted
It is relatively broad, but extraction effect largely can be affected the spectral signature with different type atural object by image segmentation precision
And textural characteristics etc. affect, the robustness of algorithm is limited by larger.Machine learning method, as emerging model training
Instrument, in landslide extraction field good effect is had been achieved for.But machine learning method generally needs substantial amounts of training sample
This, and higher requirement is distributed with to sample data.Which greatly limits machine learning method to instruct based on a scape remote sensing image
Service efficiency and practicality of the experienced model in other remote sensing images.
Additionally, the scope of 5 ' x5 ' is not only covered mostly for the research area that landslide is extracted, and Landslides are simpler
Single, background atural object mostly is vegetation, and extraction difficulty is less, algorithm less to large-scale research area and complex background atural object case study
Practicality it is in urgent need to be improved.
Present invention utilizes significance thought, the remote sensing image and DEM (Digital based on 30 meters of resolution
Elevation Model) data, it is proposed that a kind of landslide extracting method for remotely-sensed data on a large scale.Using remote sensing image
Spectral band feature by calculating the significance probability figure of image, it is by the potential extracted region on landslide out, and high with reference to DEM
Journey information, improves landslide extraction accuracy.
The content of the invention：
The purpose of the present invention is the remote sensing image for large scale, there is provided a kind of extracting method that fast and accurately comes down.Should
Method employs " significance " concept, i.e. entire image and easily causes the region that visual perception notes.By choosing suitable ripple
Section image, makes landslide areas have higher gray value relative to background atural object, it is believed that to be salient region, and then using aobvious
The method of work property extracted region extracts landslide.In landslide disaster, particularly great landslide disaster can be to the complicated back of the body for the present invention
The landslide occurred under scape atural object is efficiently monitored, and obtains generation area of coming down, so as to assess and emergency response after calamity for Disaster degree
Data supporting is provided.
To reach above-mentioned purpose, the technical solution of the present invention is：
The first step：For studying 30 meters that area chooses a scape Landsat8 images (covering 2 ° x2 ° of space) and respective regions
The dem data of resolution is experimental data；
Second step：Landsat8 image cloud removings；
1., according to the characteristic of Landsat8 image different-wavebands, the image of the 7th wave band is chosen as the basis for extracting landslide
Data, because the 7th wave band is commonly used to do geological structure investigation, can preferably distinguish landslide and other exposed soil background atural objects,
And the gray value that exposed soil region is presented in the band image is higher than vegetation area.
2. using the strong absorption characteristic of steam of the wave band of Landsat8 images the 9th, by the 9th band image binaryzation (gray value
Pixel more than 200 is considered cloud), the mask of cloud is generated, remove the cloud in 7 band images.
3rd step：Generate significance probability figure：
With landslide areas as salient region, using FASA (A Fast, Accurate, and Size-Aware
Salient Object Detection) method calculates each pixel in remote sensing image and belongs to the probability of landslide areas, and main point
For two steps：
1. space center and the variance of each color are calculated
(1) position vector P of each pixel is calculatediWith color vector Colori
Wherein, xiAnd yiIt is pixel PiHorizontal stroke, vertical coordinate, L* (Pi), a* (Pi) and b* (Pi) it is pixel PiIn color space
The gray value of each passage in CIEL*a*b*, CIEL*a*b* color spaces are usually used in image segmentation and color quantizing.
(2) each pixel P is calculatediIn space center (M both horizontally and verticallyx, My) and color variance (Vx, Vy), it is
The pixel region of high variance is strengthened below is prepared
Wherein, Mx(Pi) and Vx(Pi) pixel P is represented respectivelyiSpace center in the horizontal direction and color variance, vertically
Space center and color variance on direction can adopt similar formula to calculate.Color weight wc(Colori, Colorj) can be with
Calculated by Gaussian function：
(3) color in image is re-quantized to into Nc kind colors according to histogram distribution, calculates the space of each color
Center and color variance：
Wherein, QckRepresent the kth kind color after quantifying, hjRepresent individual for the pixel of jth kind color by i-th kind of color quantizing
Number.
2. the probability that each pixel in image belongs to significance object is calculated
Pixel PiThe probability for belonging to potential region of coming down is：
Wherein, nwAnd nhThe width and height of difference representative image, coefficient μ and ∑ are respectively：
4th step：Exposed soil background atural object is removed using morphological method
1. under normal circumstances, exposed soil floor space compared with landslide areas is larger, and the trifling connection of multiple specklees is presented
Form.Therefore, using morphology principle, 6 dilation operations are continuously done to significance probability figure, by exposed soil trifling in image
Speckle connection is got up, and forms big connected region.The concrete principle of dilation operation is as follows：
Wherein, f (x, y) is input picture, and b (x, y) is structural element.
Because significance probability figure describes the probability that pixel belongs to landslide, can be by by continuous several times dilation operation
The original larger exposed soil speckle of area is coupled together so that exposed soil integrally becomes much larger, and less, suffered shadow is taken up an area in landslide areas
Ring little.
2. calculate each connected region outsourcing rectangle wide and height, if greater than entire image wide and high ten/
One, then it is assumed that be the larger exposed soil region of floor space, corresponding region, i.e. gray value are rejected from significance probability figure and is arranged
For 0.
5th step：With reference to dem data, landslide areas are further extracted
Because landslide is mostly occurred on hillside, corresponding landslide areas gray value is higher in altitude data, by elevation map
Remove in the result images that pixel of the gray value less than or equal to 5 is all obtained from step 4 as in, obtain final landslide and extract knot
Fruit is schemed.
Description of the drawings：
Fig. 1 is flow chart provided in an embodiment of the present invention.
Fig. 2 is panorama sketch provided in an embodiment of the present invention (the 7th band image).
Fig. 3 is panorama dem data provided in an embodiment of the present invention.
Fig. 4 removes panorama sketch after cloud (the 7th band image) for provided in an embodiment of the present invention.
Fig. 5 is significance probability panorama sketch provided in an embodiment of the present invention.
Fig. 6 is the potential area results panorama sketch in landslide provided in an embodiment of the present invention.
Fig. 7 extracts result panorama sketch for landslide provided in an embodiment of the present invention.
Fig. 8 extracts the figure of result detailed example one for landslide provided in an embodiment of the present invention：A () is landslide example pseudo color coding hologram figure
(the 5th, 4,3 band combination image)；B () extracts result figure for landslide.
Fig. 9 extracts the figure of result detailed example two for landslide provided in an embodiment of the present invention：A () is landslide example pseudo color coding hologram figure
(the 5th, 4,3 band combination image)；B () extracts result figure for landslide.
Figure 10 extracts the figure of result detailed example three for landslide provided in an embodiment of the present invention：A () is landslide example pseudo color coding hologram
Figure (the 5th, 4,3 band combination image)；B () extracts result figure for landslide.
Specific embodiment：
The technical scheme in the embodiment of the present application is described below in conjunction with the accompanying drawing in the embodiment of the present application.It is aobvious
So, described embodiment is only the section Example of the application, is not all of example.
Embodiments herein with Nepal near Himalaya region as study area, choose 1 day 30 June in 2015
The one scape Landsat8 images (cover 2 ° x2 ° of space) (as shown in Figure 2) of rice resolution and 30 meters of resolution of respective regions
Dem data is experimental data (as shown in Figure 3).Cloud shown in Fig. 2 can utilize the 9th band image of Landsat8 to generate cloud and cover
Film, and then be removed, obtain Fig. 4.
As shown in figure 5, FASA methods can preferably strengthen exposed soil area information, weaken the information such as vegetation.But, exposed soil
In have major part to belong to the non-landslide areas such as construction land, and trifling big speckle shape is presented；It is tiny, trifling that landslide is presented
Shape.In order to preferably distinguish non-landslide areas and landslide areas in exposed soil, caused greatly using the expanding method in morphology operations
The trifling exposed soil region of speckle is interconnected, and by contrast, landslide areas still floor space is less.Therefore, by contrast
The size of the boundary rectangle of connected region and image after dilation operation, by the outer of wide and tall and big wide and high 1/10th in image
The connected region for connecing rectangle is rejected, and remaining connected region is the preliminary potential region in landslide extracted, as shown in Figure 6.Due to landslide
Mostly occur where the physical features such as hillside are higher, the DEM elevation maps with reference to shown in Fig. 3 all go pixel of the gray value less than 5
Remove, obtain final landslide and extract result Fig. 7.
In order in more detail, clearly illustrate the performance that the application is extracted on landslide, landslide Typical Areas conduct at three has been intercepted
Embodiment (as Figure 8-Figure 10), each of which embodiment all includes the 5th, 4 and 3 band combination by Landsat8
Pseudo color coding hologram figure and a landslide extract result figure, pseudo color coding hologram figure be display in order to become apparent from vegetation mainly to carry on the back
The landslide areas of scape.Landslide in Fig. 8-Figure 10 is preferably extracted, with certain application potential.
Claims (6)
1. it is a kind of based on remote sensing image on a large scale and the landslide extracting method of altitude data, it is characterised in that the method is for big
Area comes down, and implementation process includes that remote sensing image cloud removing, salient region strengthen, morphological operation extracts connected region and knot
Close elevation information and extract landslide, concrete steps operation is as follows：
(1) for study area choose 30 meters of resolution of a scape multispectral Landsat8 remote sensing images (cover 2 ° x2 ° of space) and
The altitude data of 30 meters of resolution of respective regions is experimental data；
(2) Landsat8 images cloud removing：
According to the characteristic of Landsat8 image different-wavebands, the basic data of the image as extraction landslide of the 7th wave band is chosen, because
It is commonly used to do geological structure investigation for the 7th wave band, can preferably distinguishes landslide and other exposed soil background atural objects, and exposed soil
The gray value that region is presented in the band image is higher than other atural objects；
Using the strong absorption characteristic of steam of the wave band of Landsat8 images the 9th, by the 9th band image binaryzation, (gray value is more than 200
Pixel be considered cloud), generate cloud mask, remove 7 band images in cloud；
(3) significance probability figure is generated：
With landslide areas as salient region, using FASA (A Fast, Accurate, and Size-Aware Salient
Object Detection) method calculates each pixel in remote sensing image and belongs to the probability of landslide areas；
(4) exposed soil background atural object is removed using morphological method：
Under normal circumstances, the exposed soil on non-landslide floor space compared with landslide areas is larger, and it is trifling that multiple big specklees are presented
The feature of connection；Therefore, using morphology principle, 6 dilation operations are carried out continuously to significance probability figure, will be trifling in image
The connection of exposed soil speckle get up, form big connected region；
Because significance probability figure describes the probability that pixel belongs to landslide, can originally by continuous several times dilation operation
The larger exposed soil speckle of area is coupled together so that exposed soil integrally becomes much larger, and landslide areas occupation of land is less, and institute is impacted not
Greatly；
The wide and height of the outsourcing rectangle of each connected region is calculated, if greater than wide and high 1/10th of entire image, is then recognized
To be the larger exposed soil region of floor space, corresponding region is rejected from significance probability figure, its gray value is set to into 0；
(5) altitude data is combined, landslide areas are further extracted：
Because landslide is mostly occurred on hillside, corresponding landslide areas gray value is higher in altitude data, by elevation map picture
Remove in the result images that pixel of the gray value less than or equal to 5 is all obtained from step (4), obtain final landslide and extract result
Figure.
2. as claimed in claim 1 based on remote sensing image and the landslide extracting method of altitude data, it is characterised in that described
(1) step, described remote sensing image is the multispectral image of 30 meters of resolution after the calamity of single scape landslide, and present patent application is employed
Landsat8 remote sensing images.
3. as claimed in claim 1 based on remote sensing image and the landslide extracting method of altitude data, it is characterised in that described
(2) step, data based on the described band image of selection the 7th, wave band is commonly used to do geological structure investigation, and exposed soil region
Higher gray value is presented compared to other background atural objects.
4. as claimed in claim 1 based on remote sensing image and the landslide extracting method of altitude data, it is characterised in that described
(3) step, it is described based on " FASA (A Fast, Accurate, and Size-Aware Salient Object
Detection) " method generates significance probability figure, wherein：Built according to the histogram distribution of image and mapped, by the face of image
Color re-quantization is less color, and then strengthens difference of the exposed soil relative to other atural objects, calculates the color after each quantization
Space center and color center, and the probability that each pixel belongs to salient region is calculated according to gaussian kernel function.
5. as claimed in claim 1 based on remote sensing image and the landslide extracting method of altitude data, it is characterised in that described
(4) step, the dilation operation in described employing morphology couples together speckle trifling in image, and then highlights non-landslide
Exposed soil region accounts for image larger proportion this feature, by connected region attribute selection, area is less, and in compact shape
Region remains, and constitutes the potential extracted region figure in landslide.
6. as claimed in claim 1 based on remote sensing image and the landslide extracting method of altitude data, it is characterised in that described
(5) step, described combination altitude data, it is considered to which landslide generally occurs on hillside, by positioned at the relatively low potential landslide of flat, physical features
Region is further rejected, and obtains the final extraction result for coming down.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610718410.9A CN106600574B (en) | 2016-08-25 | 2016-08-25 | A kind of landslide extracting method based on remote sensing image and altitude data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610718410.9A CN106600574B (en) | 2016-08-25 | 2016-08-25 | A kind of landslide extracting method based on remote sensing image and altitude data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106600574A true CN106600574A (en) | 2017-04-26 |
| CN106600574B CN106600574B (en) | 2019-07-23 |
Family
ID=58555692
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610718410.9A Expired - Fee Related CN106600574B (en) | 2016-08-25 | 2016-08-25 | A kind of landslide extracting method based on remote sensing image and altitude data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106600574B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108280812A (en) * | 2018-01-23 | 2018-07-13 | 中国科学院遥感与数字地球研究所 | A kind of excessive fire method for extracting region based on image enhancement |
| CN108647567A (en) * | 2018-03-29 | 2018-10-12 | 中国人民解放军61540部队 | Scene identifiability analysis method based on Conditional Evidence theory |
| CN108846347A (en) * | 2018-06-06 | 2018-11-20 | 广西师范学院 | A kind of rapid extracting method in highway landslide region |
| CN109767409A (en) * | 2018-11-28 | 2019-05-17 | 中国科学院遥感与数字地球研究所 | Landslide change detecting method, storage medium and electronic equipment based on remote sensing image |
| CN110060273A (en) * | 2019-04-16 | 2019-07-26 | 湖北省水利水电科学研究院 | Remote sensing image landslide plotting method based on deep neural network |
| CN111028255A (en) * | 2018-10-10 | 2020-04-17 | 千寻位置网络有限公司 | Farmland area pre-screening method and device based on prior information and deep learning |
| CN111104896A (en) * | 2019-12-18 | 2020-05-05 | 河南大学 | Automatic land surface water identification method based on Sentinel-1 microwave satellite image |
| CN111160296A (en) * | 2019-12-31 | 2020-05-15 | 中国科学院电子学研究所 | Landslide disaster detection method |
| CN111626269A (en) * | 2020-07-07 | 2020-09-04 | 中国科学院空天信息创新研究院 | Practical large-space-range landslide extraction method |
| CN112834432A (en) * | 2021-01-08 | 2021-05-25 | 兰州大学 | Landslide thickness inversion method based on remote sensing technology and kinematics principle |
| CN114299290A (en) * | 2021-12-24 | 2022-04-08 | 腾晖科技建筑智能(深圳)有限公司 | Bare soil identification method, device, equipment and computer readable storage medium |
| CN116108758A (en) * | 2023-04-10 | 2023-05-12 | 中南大学 | Landslide susceptibility evaluation method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254174A (en) * | 2011-07-08 | 2011-11-23 | 中铁第四勘察设计院集团有限公司 | Method for automatically extracting information of bare area in slumped mass |
| CN102426395A (en) * | 2011-11-11 | 2012-04-25 | 铁道第三勘察设计院集团有限公司 | Slip mass volume parameter measuring method based on three-dimensional geologic remote sensing interpretation technology |
| CN104700399A (en) * | 2015-01-08 | 2015-06-10 | 东北大学 | Method for demarcating large-deformation landslide displacement field based on high-resolution remote sensing image |
| CN104899437A (en) * | 2015-05-29 | 2015-09-09 | 杭州辰青和业科技有限公司 | Early-warning method for heavy-rainfall type landslide hazard |
| KR20160061445A (en) * | 2014-11-17 | 2016-06-01 | 서울대학교산학협력단 | Method and System of Construction of Landslide Hazard Map During Earthquakes Considering Geometrical Amplification Characteristics of Slope |
-
2016
- 2016-08-25 CN CN201610718410.9A patent/CN106600574B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254174A (en) * | 2011-07-08 | 2011-11-23 | 中铁第四勘察设计院集团有限公司 | Method for automatically extracting information of bare area in slumped mass |
| CN102426395A (en) * | 2011-11-11 | 2012-04-25 | 铁道第三勘察设计院集团有限公司 | Slip mass volume parameter measuring method based on three-dimensional geologic remote sensing interpretation technology |
| KR20160061445A (en) * | 2014-11-17 | 2016-06-01 | 서울대학교산학협력단 | Method and System of Construction of Landslide Hazard Map During Earthquakes Considering Geometrical Amplification Characteristics of Slope |
| CN104700399A (en) * | 2015-01-08 | 2015-06-10 | 东北大学 | Method for demarcating large-deformation landslide displacement field based on high-resolution remote sensing image |
| CN104899437A (en) * | 2015-05-29 | 2015-09-09 | 杭州辰青和业科技有限公司 | Early-warning method for heavy-rainfall type landslide hazard |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108280812B (en) * | 2018-01-23 | 2022-03-11 | 中国科学院遥感与数字地球研究所 | Image enhancement-based fire passing area extraction method |
| CN108280812A (en) * | 2018-01-23 | 2018-07-13 | 中国科学院遥感与数字地球研究所 | A kind of excessive fire method for extracting region based on image enhancement |
| CN108647567A (en) * | 2018-03-29 | 2018-10-12 | 中国人民解放军61540部队 | Scene identifiability analysis method based on Conditional Evidence theory |
| CN108647567B (en) * | 2018-03-29 | 2021-10-29 | 中国人民解放军61540部队 | Scene identifiability analysis method based on condition evidence theory |
| CN108846347B (en) * | 2018-06-06 | 2021-06-08 | 广西师范学院 | Rapid extraction method for road landslide area |
| CN108846347A (en) * | 2018-06-06 | 2018-11-20 | 广西师范学院 | A kind of rapid extracting method in highway landslide region |
| CN111028255A (en) * | 2018-10-10 | 2020-04-17 | 千寻位置网络有限公司 | Farmland area pre-screening method and device based on prior information and deep learning |
| CN111028255B (en) * | 2018-10-10 | 2023-07-21 | 千寻位置网络有限公司 | Farmland area pre-screening method and device based on priori information and deep learning |
| CN109767409A (en) * | 2018-11-28 | 2019-05-17 | 中国科学院遥感与数字地球研究所 | Landslide change detecting method, storage medium and electronic equipment based on remote sensing image |
| CN109767409B (en) * | 2018-11-28 | 2021-04-09 | 中国科学院空天信息创新研究院 | Landslide change detection method based on remote sensing image, storage medium and electronic equipment |
| CN110060273B (en) * | 2019-04-16 | 2021-05-18 | 湖北省水利水电科学研究院 | Remote sensing image landslide mapping method based on deep neural network |
| CN110060273A (en) * | 2019-04-16 | 2019-07-26 | 湖北省水利水电科学研究院 | Remote sensing image landslide plotting method based on deep neural network |
| CN111104896A (en) * | 2019-12-18 | 2020-05-05 | 河南大学 | Automatic land surface water identification method based on Sentinel-1 microwave satellite image |
| CN111104896B (en) * | 2019-12-18 | 2021-08-27 | 河南大学 | Automatic land surface water identification method based on Sentinel-1 microwave satellite image |
| CN111160296A (en) * | 2019-12-31 | 2020-05-15 | 中国科学院电子学研究所 | Landslide disaster detection method |
| CN111160296B (en) * | 2019-12-31 | 2024-01-26 | 中国科学院电子学研究所 | Landslide hazard detection method |
| CN111626269B (en) * | 2020-07-07 | 2021-08-27 | 中国科学院空天信息创新研究院 | Practical large-space-range landslide extraction method |
| CN111626269A (en) * | 2020-07-07 | 2020-09-04 | 中国科学院空天信息创新研究院 | Practical large-space-range landslide extraction method |
| CN112834432A (en) * | 2021-01-08 | 2021-05-25 | 兰州大学 | Landslide thickness inversion method based on remote sensing technology and kinematics principle |
| CN114299290A (en) * | 2021-12-24 | 2022-04-08 | 腾晖科技建筑智能(深圳)有限公司 | Bare soil identification method, device, equipment and computer readable storage medium |
| CN114299290B (en) * | 2021-12-24 | 2023-04-07 | 腾晖科技建筑智能(深圳)有限公司 | Bare soil identification method, device, equipment and computer readable storage medium |
| CN116108758A (en) * | 2023-04-10 | 2023-05-12 | 中南大学 | Landslide susceptibility evaluation method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106600574B (en) | 2019-07-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106600574A (en) | Landslide extraction method based on remote-sensing image and altitude data | |
| Li et al. | TransUNetCD: A hybrid transformer network for change detection in optical remote-sensing images | |
| CN110929607B (en) | Remote sensing identification method and system for urban building construction progress | |
| Li et al. | Principal component analysis of stacked multi-temporal images for the monitoring of rapid urban expansion in the Pearl River Delta | |
| CN108629287A (en) | A kind of remote sensing image terrain classification method | |
| Gao et al. | Multitemporal Landsat missing data recovery based on tempo-spectral angle model | |
| CN109308451A (en) | A kind of high score data information extraction system and method | |
| Lv et al. | ShapeFormer: A shape-enhanced vision transformer model for optical remote sensing image landslide detection | |
| CN110647909A (en) | Remote sensing image classification method based on three-dimensional dense convolution neural network | |
| CN113221813A (en) | Coastline remote sensing extraction method | |
| Yang et al. | Object-based classification of cloudy coastal areas using medium-resolution optical and SAR images for vulnerability assessment of marine disaster | |
| Dong et al. | A review of research on remote sensing images shadow detection and application to building extraction | |
| Wang et al. | Edge enhanced channel attention-based graph convolution network for scene classification of complex landscapes | |
| Liu et al. | Intelligent identification of landslides in loess areas based on the improved YOLO algorithm: a case study of loess landslides in Baoji City | |
| CN106960119A (en) | A kind of fusion DI forest disturbance comprehensive characteristics index and method | |
| Wang et al. | Rapid identification of post-earthquake collapsed buildings via multi-scale morphological profiles with multi-structuring elements | |
| Hao et al. | A subpixel mapping method for urban land use by reducing shadow effects | |
| Wang et al. | Utilizing parallel networks to produce sub-pixel shifted images with multiscale spatio-spectral information for soft-then-hard sub-pixel mapping | |
| CN115497004A (en) | Intelligent detection and extraction method for hidden danger of railway external environment based on remote sensing data | |
| Li et al. | Subpixel change detection based on improved abundance values for remote sensing images | |
| Buehler et al. | Rapid mapping with remote sensing data during flooding 2005 in Switzerland by object-based methods: A case study | |
| Miao et al. | Deep learning landslide extraction based on multi-temporal image difference local spatial information | |
| CN106339423A (en) | Method and device for dynamically updating sugarcane planting information | |
| Yeh et al. | Principal component analysis of stacked multi-temporal images for the monitoring of rapid urban expansion in the Pearl River Delta | |
| Javed et al. | Development of Soil-Suppressed Impervious Surface Area Index for Automatic Urban Mapping |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190723 Termination date: 20200825 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |